|Publication number||US7170886 B1|
|Application number||US 09/843,596|
|Publication date||Jan 30, 2007|
|Filing date||Apr 26, 2001|
|Priority date||Apr 26, 2001|
|Also published as||US7453828|
|Publication number||09843596, 843596, US 7170886 B1, US 7170886B1, US-B1-7170886, US7170886 B1, US7170886B1|
|Inventors||Luke K. Surazski, Pascal Huart, Michael E. Knappe|
|Original Assignee||Cisco Technology, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (11), Non-Patent Citations (1), Referenced by (7), Classifications (13), Legal Events (3)|
|External Links: USPTO, USPTO Assignment, Espacenet|
1. Field of the Invention
The present invention is related to the field of network telephony, and more specifically to devices, softwares and methods for indexing telephone conversations while they are being recorded.
2. Description of the Related Art
Networks, such as Local Area Networks (LANs), Wide Area Networks (WANs), or the internet are increasingly being used for telephony. Network telephony is different from conventional telephony in many aspects. In network telephony the voice is encoded digitally, and transmitted as packets. The packets find their way through the network by complying with the so-called Internet Protocol (IP).
A problem in telephony is in recording conversations. These would require a massive data storage, since encoding voice consumes large amounts of bandwidth.
The problem is that needles of precious voice data are often hidden in haystacks of whole conversations. For example, in a telephone call with a stockbroker, a client's order forms a legally binding request. In the event of a later dispute, the client's voice placing the order must be retrieved. But that short duration of the telephone call may be buried in a long telephone conversation. For retrieving that short duration, it is necessary to play back the entire telephone call, which is time consuming. For another example, in a conference call, it may be desirable to isolate the instances where only a certain party speaks.
The present invention overcomes these problems and limitations of the prior art.
Generally, the present invention provides devices, softwares and methods for generating, in real time, indexing metadata for select portions of a telephone conversation or conference. The indexing metadata is generated responsive to inputs received while the conversation is being recorded live. The inputs may be either by a user pressing a soft key on a telephone device, or by a voice conference bridge determining who are the dominant speakers in a multi-party conference.
The invention offers an advantage when the indexing metadata is used to retrieve the select portions of the telephone conversation. The advantage is that the indexing metadata may be used to skip directly to the corresponding select portions of the telephone conversation, without having to play out the entire conversation.
The invention will become more readily apparent from the following Detailed Description, which proceeds with reference to the drawings, in which:
As has been mentioned, the present invention provides devices, softwares and methods for generating indexing metadata for select portions of a telephone conversation in real time. The conversation is between two or more parties, and at least some of it takes place through a packet switched network. The invention is now described in more detail.
Referring now to
A telephone 115 may establish a connection 119 with voice gateway 120 made according to an embodiment of the invention. It will be appreciated that a voice gateway 120 is used, because telephone 115 is an ordinary telephone. Connection 119 is therefore made through the Public Switched Telephone Network (PSTN). Equivalently, if an IP telephone were used, then voice gateway 120 might be any network switch.
Telephone 115, with the aid of voice gateway 120, may establish a communication link 132 with another voice gateway 140. Voice gateway 140 may establish a connection 143 with telephone 146. Accordingly, a user of telephone 115 may have a conversation with a user of telephone 146 through a combined communication link that includes connection 119, communication link 132 and connection 143.
The voice conversation is encoded as digital data that passes through the combined communication link. When passing through communication link 132, it is in the form of data packets 134, 136, 138. These data packets 134, 136, 138 may be encoded under a Voice over Internet Protocol (VoIP).
In the embodiment of
It will be appreciated that memory 150 is preferably implemented by a memory that has high capacity, since large amounts of digital data are typically required to represent voice. Accordingly, memory 150 is often called large memory or mass memory, to be distinguished from memory registers and incidental memories associated with processors.
Groups 164, 166, 168 may be derived from data that arrives from both telephone 115 and telephone 146. When storing voice that was generated from telephone 146, groups 164, 166, 168 may be derived from data packets 134, 136, 138 respectively. Derivation may be in a number of ways, such as by stripping from data packets 134, 136, 138 headers that facilitate navigation through IP Cloud 110. That is because such headers have fulfilled their mission by the time their encapsulated data have reached network switch 120.
Importantly according to the invention, a data group 172 is generated in real time. This means that data group 172 is generated while the conversation is taking place.
Data group 172 does not represent voice data. Data group 172 is metadata, in other words data about the voice represented by the associated voice data groups 164, 166, 168. Data group 172 indexes at least one of the associated voice data groups 164, 166, 168, for later retrieval.
Data group 172 is stored in a memory. It may stored in the same memory 150 as voice data groups 164, 166, 168, although the invention is not limited in this regard.
Referring now to
Switch 220 has a network interface 230 for interfacing with a network, such as network 110.
Switch 220 also has a processor 240 coupled with network interface 230. Processor 240 may be implemented as a Digital Signal Processor (DSP), Central Processing Unit (CPU), or any other equivalent way known in the art. Processor 240 may include a voice codec 250, which is made from an encoder 253 and a decoder 256. In other embodiments, codec 250 is not a voice codec.
Switch 220 additionally includes a memory 260, on which a program 270 may reside. Functions of processor 240 may be controlled by program 270, as will become apparent from the below.
Referring now to
In memory 350, groups of voice data 322 are arranged. The arrangement may be in a serial fashion, to reflect the order of the represented conversation. In addition, incoming data groups 172, 164, 166, 168 are stored. In this embodiment, metadata 172 is stored along with voice data groups 164, 166, 168. It is also of interest that, in the embodiment of
Referring now to
IP telephone 410 includes a handset that has a microphone 411 for receiving local voice inputs, and converting them into an analog signal. It also has a speaker 412 for converting remotely received voice data into sound, after some suitable processing of the type that is known in the art. The handset may be provided as part of a device 415 that appears as an ordinary telephone.
Device 415 includes a keyboard with keys for dialing telephone numbers, such as an access number for telephone 446. One such key 416 may produce a DTMF sound, and may be a number key or a special function key. Usually each key produces a different DTMF sound.
IP telephone 410 additionally includes a signal processing unit 417. Unit 417 may include an Analog to Digital Converter for converting the output of microphone 411 into local digital voice data. Unit 417 may also include a Digital to Analog Converter for converting remote digital voice data into an analog signal suitable for speaker 412.
Unit 417 may also include a network interface for establishing an IP connection 419 with a network switch 420 in network 110. Data is exchanged through connection 419. In fact, network switch 420 may establish a connection 432 with a voice gateway 440 through network 110. And voice gateway 440 may establish a connection 443 with an ordinary telephone 446. This way users of ordinary telephone 446 and IP telephone 410 may have a conversation.
Data is received through connection 432 in the form of data packets 434, 436, 438. These contain remote digital voice data, which encode voice generated at the ordinary telephone 446.
Unit 417 receives the remote digital voice data through switch 420 and link 419. In addition, unit 417 may store portions of the aggregate local and remote digital voice data in a memory 450. In the embodiment of
Unit 417 generates indexing metadata about a select portion of the aggregate voice data in response to a tagging input. In the embodiment of
In the embodiment of
Referring now to
In the embodiment of
The functions of the keys 416-A, 416-B, 416-C may be conveyed in a number of ways. One such way is to have keys 416-A, 416-B, 416-C be hard keys, which bear proper inscriptions. Another such way is for them to be soft keys, such as labels on icons of keys on a screen. The labels may be the same, or changing dynamically, depending on the flow of the conversation. The soft keys may be selected by a pointing device such as a mouse or a “Tab” key, and activated by a mouse button or an “Enter” key. In effect, soft keys are programmable by the device. Their behavior is configurable.
In one embodiment, tagging lasts while each of three keys 416-A, 416-B, 416-C is pressed. So, when key 416-A is pressed, a time T1 is recorded, under a suitable variable name NAME_START. When key 416-A is released, a time T2 is recorded, under a suitable variable name NAME_END. Similarly with recording the times T3, T4 for key 416-B, and times T5, T6 for key 416-C.
In another embodiment, each of three keys 416-A, 416-B, 416-C is to be pressed and released. The first pressing indicates T1, T3, T5, respectively, while the second pressing indicates T2, T4, T6, respectively.
Referring now to
In the embodiment of
Referring now to
Bridge 810 is for placing in a network, such as IP Cloud 110. Bridge 810 is adapted to establish communication links through network 110. More specifically, a connection 812 is established with voice gateway 140, a connection 814 is established with voice gateway 440, and a connection 816 is established with network switch 840. These three links are concurrent in that they are used at the same time, even though they may not be established at the same instant, as is known in the art.
IP phone 847 establishes a connection 843 with network switch 840. This way, users of telephones 146, 446 and 843 are capable of having a simultaneous conference, through bridge 810.
It should be noted that, while all three links 812, 814, 816 are through network 110, that is not necessary for practicing the invention. Indeed, one of them may be through an ordinary telephone line. Also, bridge 810 may be located in the same place as one of the speakers, in which case the corresponding link need not be through a network at all.
Bridge 810 receives data from all three communication links 812, 814, 816, which it considers as channels. In the instant depicted in
There are a number of features that conference bridge 810 may have.
Nominally, bridge 810 sums all the received inputs, and plays them back through the channels. This way all users hear an aggregate of what is said, as if they were present in the same location.
As a first improvement, bridge 810 does not play to each user what that user said. This way, each user hears an aggregate of only what the others said.
As a second improvement, bridge 810 continuously chooses to favor at least one of the channels at the expense of disfavoring at least another one of the channels. The selection is performed by a channel selector.
The channel selector analyzes continuously the inputs from the channels, in trying to determine which one(s) are the dominant one(s). The determination may be by which channel is the loudest. And as the users speak at different times, the channel selector changes which channel(s) it favors. It may deselect the favored channel, and favor another one. This way, at any given time, each user hears only what bridge 810 deems is the important part of a voice conference.
In the example of
In the embodiment of
Memory 850 is large, as discussed above. Memory 850 may be collocated with bridge 810 in some embodiments. In other embodiments, memory 850 may be accessed by establishing a storage connection 818 with a further network switch 870, concurrently with connections 812, 814, 186. Network switch 870 receives voice data packets 862, 864, 866, over storage connection 818, and forwards them to memory 850. These may be further stripped from the headers and trailers for more economical storage.
Importantly according to the invention, bridge 810 generates a selection identifier about the selection of which communication link to favor, and which to disfavor at any given time. The selection identifier may be encoded as a data group, which in turn may be encoded as a packet 872. The selection identifier may include an identification code for the party whose channel is favored, and a time stamp of when the selection is made. The time stamp may be an RTP time stamp, which may then be converted into a different format. Such a different format may advantageously be standard user-readable format for retrieval while playback. Such a standard may be one of the SMPTE (Society of Motion Picture and TV Engineers) standards.
Bridge 810 then stores the selection identifier. This may be accomplished by transmitting data group 872 to memory 850. Transmitting may be accomplished through storage connection 818. Alternately, the selection identifier may be stored in a different memory.
Referring now to
Bridge 910 has a network interface 930 for interfacing with a network, such as network 110.
Bridge 910 also has a processor 940 coupled with network interface 930. Processor 940 may be implemented as a Digital Signal Processor (DSP), Central Processing Unit (CPU), or any other equivalent way known in the art. Processor 940 may include a codec 950, which is made from an encoder 953 and a decoder 956. Codec 950 may or may not be a voice codec.
Bridge 910 moreover includes a channel selector 958, which may be implemented as part of processor 940. Channel selector 958 determines which channel(s) will be favored over which channel(s). The input from the disfavored ones is ignored. The inputs from the favored ones are added. Channel selector 958 may make the determination by employing a speaker selection algorithm, which determines who it regards as the dominant speakers.
Bridge 910 additionally includes a memory 960, on which a program 970 may reside. Functions of processor 940 may be controlled by program 970, as will become apparent from the below.
It is readily apparent that the present invention may be implemented by one or more devices that include logic circuitry. It may also be implemented by a device that includes a dedicated processor system, which may include a microcontroller or a microprocessor.
The invention additionally provides methods, which are described below. Moreover, the invention provides apparatus that performs, or assists in performing the methods of the invention. This apparatus may be specially constructed for the required purposes, or it may comprise a general-purpose computer selectively activated or reconfigured by a computer program stored in the computer. The methods and algorithms presented herein are not necessarily inherently related to any particular computer or other apparatus. In particular, various general-purpose machines may be used with programs in accordance with the teachings herein, or it may prove more convenient to construct more specialized apparatus to perform the required method steps. The required structure for a variety of these machines will appear from this description.
Useful machines or articles for performing the operations of the present invention include general-purpose digital computers or other similar devices. In all cases, there should be borne in mind the distinction between the method of operating a computer and the method of computation itself. The present invention relates also to method steps for operating a computer and for processing electrical or other physical signals to generate other desired physical signals.
The invention additionally provides a program, and a method of operation of the program. The program is most advantageously implemented as a program for a computing machine, such as a general-purpose computer, a special purpose computer, a microprocessor, etc.
The invention also provides a storage medium that has the program of the invention stored thereon. The storage medium is a computer-readable medium, such as a memory, and is read by the computing machine mentioned above.
A program is generally defined as a sequence of steps leading to a desired result. These steps, also known as instructions, are those requiring physical manipulations of physical quantities. Usually, though not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated or processed. When stored, they may be stored in any computer-readable medium. It is convenient at times, principally for reasons of common usage, to refer to these signals as bits, data bits, samples, values, elements, symbols, characters, images, terms, numbers, or the like. It should be borne in mind, however, that all of these and similar terms are associated with the appropriate physical quantities, and that these terms are merely convenient labels applied to these physical quantities.
This detailed description is presented largely in terms of flowcharts, display images, algorithms, and symbolic representations of operations of data bits within a computer readable medium, such as a memory. Such descriptions and representations are the type of convenient labels used by those skilled in programming and/or the data processing arts to effectively convey the substance of their work to others skilled in the art. A person skilled in the art of programming may use this description to readily generate specific instructions for implementing a program according to the present invention. For the sake of economy, however, flowcharts used to describe methods of the invention are not repeated in this document for describing software according to the invention.
Often, for the sake of convenience only, it is preferred to implement and describe a program as various interconnected distinct software modules or features, collectively also known as software. This is not necessary, however, and there may be cases where modules are equivalently aggregated into a single program with unclear boundaries. In any event, the software modules or features of the present invention may be implemented by themselves, or in combination with others. Even though it is said that the program may be stored in a computer-readable medium, it should be clear to a person skilled in the art that it need not be a single memory, or even a single machine. Various portions, modules or features of it may reside in separate memories, or even separate machines. The separate machines may be connected directly, or through a network, such as a local access network (LAN), or a global network, such as the Internet.
In the present case, methods of the invention are implemented by machine operations. In other words, embodiments of the program of the invention are made such that they perform methods of the invention that are described in this document. These may be optionally performed in conjunction with one or more human operators performing some, but not all of them. As per the above, the users need not be collocated with each other, but each only with a machine that houses a portion of the program. Alternately, some of these machines may operate automatically, without users and/or independently from each other.
Methods of the invention are now described.
Referring now to
According to a box 1110, a communication link is established through a network. The link will be used for receiving voice data in the form of packets. The voice data will come in portions, some of which will be tagged.
According to a next box 1115, it is inquired whether a tagging input has been received.
If not, then according to a next box 1140, a next portion of voice data is received from the communication link.
If yes, then according to a next box 1150, a tag is generated about the portion. Then according to an optional next box 1160, the tag is stored, and execution continues to box 1140.
After box 1140, according to an optional next box 1170, the received portion of voice data is converted into sound.
According to an optional next box 1180, it is again whether a tag was generated. This is substantially the same question as was asked in step 1115. If not, then execution may return to box 1115, to ask whether a next tagging input was received.
If yes, then according to a next box 1190, the received portion of voice data is stored in a large memory. Execution then returns to box 1115.
Accordingly, if box 1180 has taken place, then only the tagged portions are thus stored. If box 1180 does not take place, then all the portions are stored, whether tagged or not.
Referring now to
According to a box 1210, a communication link is established through a network. The link will be used for receiving voice data in the form of packets. The voice data will come in portions, some of which will be tagged.
According to a next box 1215, it is inquired whether a tagging input has been received. As per the above, the tagging input may be a DTMF signal, etc. If multiple indexing features are enabled (e.g. one for “NAME”, one for “ACCOUNT #”, etc.), then it is not yet known which one was activated.
If yes, then according to a next box 1220, it is inquired whether a first indexing feature was activated by the tagging input. If yes, then according to a next box 1223, a first tag is generated about the voice data. According to a next box 1226, the first tag is stored.
If, at box 1220, it is determined that the first feature was not activated, then, according to a next box 1230, it is inquired whether a second indexing feature was activated by the tagging input. If yes, then according to a next box 1233, a second tag is generated about the voice data. According to a next box 1236, the second tag is stored.
According to a next box 1240, a portion of the next voice data is received. This takes place if no tagging input is received at box 1215, or after box 1226, or after box 1236.
According to an optional next box 1250, the received portion of voice data is converted into sound.
According to a next box 1260, the received portion of voice data is stored in a large memory. Execution then returns to box 1215.
Referring now to
According to a box 1310, at least three communication links are established. Of those, at least the first and the second are through a network.
According to an optional next box 1320, three streams of voice data are received from the three communication links, one from each. Sometimes, the voice data may contain zero voice, or be a Silence Indication Packet.
According to an optional next box 1330, the first communication link is selected for favoring, while the second communication link is selected for disfavoring. The determination may be from the content of the streams of voice data.
According to an optional next box 1340, it is inquired whether this was the first selection.
If yes, then according to a next box 1350, a selection identifier is generated and stored. The selection identifier may identify the party of the first communication link. It may also include an RTP time stamp. The RTP time stamp may then be converted to a different format, such as an SMPTE format.
If this were not the first selection then according to a next box 1360, it is inquired whether the selection is different from the immediately previous one. If yes, then the new selection includes a deselection of the previous selection. According to an optional next box 1370, a deselection identifier is generated and stored. It may include an RTP time stamp, which may then be converted to a different format. Execution then proceeds to box 1350.
If, at box 1360, the selection is not different from the previous one, or after box 1350, then, according to an optional next box 1380, the first but not the second voice data is retransmitted through the third communication link, and also optionally through the second communication link. The transmission of box 1380 may also include voice data from yet another speaker who is favored over the second voice data.
According to an optional next box 1390, the first but not the second voice data is stored in a large memory. Execution then returns to box 1320.
The devices, softwares and methods of the present invention are used for real time recording. The indexing tags of the invention make it easy to scan a recorded conversation or conference for items of interest. The indexing tags are used by a playback device to skip to either an indexed portion of a conversation, or to instances of speech of a desired speaker. In a generic application, the playback device would include a “PREVIOUS INDEX” and a “NEXT INDEX” buttons.
A person skilled in the art will be able to practice the present invention in view of the description present in this document, which is to be taken as a whole. Numerous details have been set forth in order to provide a more thorough understanding of the invention. In other instances, well-known features have not been described in detail in order not to obscure unnecessarily the invention.
While the invention has been disclosed in its preferred form, the specific embodiments as disclosed and illustrated herein are not to be considered in a limiting sense. Indeed, it should be readily apparent to those skilled in the art in view of the present description that the invention may be modified in numerous ways. The inventor regards the subject matter of the invention to include all combinations and subcombinations of the various elements, features, functions and/or properties disclosed herein.
The following claims define certain combinations and subcombinations, which are regarded as novel and non-obvious. Additional claims for other combinations and subcombinations of features, functions, elements and/or properties may be presented in this or a related document.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5883891 *||Apr 30, 1996||Mar 16, 1999||Williams; Wyatt||Method and apparatus for increased quality of voice transmission over the internet|
|US6377995 *||Feb 19, 1998||Apr 23, 2002||At&T Corp.||Indexing multimedia communications|
|US6542499 *||Jul 9, 2001||Apr 1, 2003||Cisco Technology, Inc.||PSTN fallback using dial on demand routing scheme|
|US6611803 *||Dec 14, 1999||Aug 26, 2003||Matsushita Electric Industrial Co., Ltd.||Method and apparatus for retrieving a video and audio scene using an index generated by speech recognition|
|US6751297 *||Dec 11, 2000||Jun 15, 2004||Comverse Infosys Inc.||Method and system for multimedia network based data acquisition, recording and distribution|
|US6781962 *||Apr 26, 2002||Aug 24, 2004||Jetque||Apparatus and method for voice message control|
|US6833865 *||Jul 29, 1999||Dec 21, 2004||Virage, Inc.||Embedded metadata engines in digital capture devices|
|US6847598 *||Mar 9, 2001||Jan 25, 2005||Fujitsu Limited||Servo with digital filter to control gain in a frequency band where open loop characteristic is higher than the phase cross-over frequency and lower than resonance frequency|
|US6857130 *||Apr 6, 2001||Feb 15, 2005||Sun Microsystems, Inc.||Resynchronizing media during streaming|
|US6887641 *||Nov 19, 2002||May 3, 2005||Eastman Kodak Company||Mammography imaging method using high peak voltage and rhodium or tungsten anodes|
|US6947598 *||Apr 20, 2001||Sep 20, 2005||Front Porch Digital Inc.||Methods and apparatus for generating, including and using information relating to archived audio/video data|
|1||Lenny Grant, Finding needles in Video Haystacks, Technology Investor, Jan. 2001 pp. 28-30.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US8379819 *||Dec 24, 2008||Feb 19, 2013||Avaya Inc||Indexing recordings of telephony sessions|
|US8712464 *||Jan 18, 2011||Apr 29, 2014||Telefonaktiebolaget L M Ericsson (Publ)||Method and apparatus for communication channel error rate estimation|
|US9160551||Mar 23, 2012||Oct 13, 2015||Cisco Technology, Inc.||Analytic recording of conference sessions|
|US9235475 *||Mar 5, 2013||Jan 12, 2016||Emc Corporation||Metadata optimization for network replication using representative of metadata batch|
|US20100158204 *||Dec 24, 2008||Jun 24, 2010||Alan Diskin||Indexing recordings of telephony sessions|
|US20110111790 *||Jan 18, 2011||May 12, 2011||Lennart Andersson||Method and Apparatus for Communication Channel Error Rate Estimation|
|WO2008148930A1 *||Jun 5, 2008||Dec 11, 2008||Nokia Corporation||System and method for storing multiparty vtoeo conferencing presentations|
|U.S. Classification||370/352, 370/401, 370/392|
|International Classification||H04L12/66, H04L12/28|
|Cooperative Classification||H04L65/1083, H04L65/403, H04L65/605, H04L29/06027|
|European Classification||H04L29/06C2, H04L29/06M6C6, H04L29/06M2S4, H04L29/06M4C|
|Apr 26, 2001||AS||Assignment|
Owner name: CISCO TECHNOLOGY, INC., CALIFORNIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:SURAZSKI, LUKE K.;HUART, PASCAL;KNAPPE, MICHAEL E.;REEL/FRAME:011757/0059;SIGNING DATES FROM 20010327 TO 20010403
|Jul 30, 2010||FPAY||Fee payment|
Year of fee payment: 4
|Jul 30, 2014||FPAY||Fee payment|
Year of fee payment: 8